Society for American Archaeology

Point Typologies, Cultural Transmission, and the Spread of Bow-and-Arrow Technology in

the Prehistoric Great Basin
Author(s): Robert L. Bettinger and Jelmer Eerkens
Source: American Antiquity, Vol. 64, No. 2 (Apr., 1999), pp. 231-242
Published by: Society for American Archaeology
Stable URL: http://www.jstor.org/stable/2694276
Accessed: 16-05-2016 13:33 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.

Society for American Archaeology is collaborating with JSTOR to digitize, preserve and extend access to
American Antiquity

This content downloaded from 144.82.199.209 on Mon, 16 May 2016 13:33:55 UTC
All use subject to http://about.jstor.org/terms
 POINT TYPOLOGIES, CULTURAL TRANSMISSION, AND THE SPREAD OF
BOW-AND-ARROW TECHNOLOGY IN THE PREHISTORIC GREAT BASIN

Robert L. Bettinger and Jelmer Eerkens

Decrease in projectile point size around 1350 B.P. is commonly regarded as marking the replacement of the atlatl by the bow
and arrow across the Great Basin. The point typology most widely employed in the Great Basin before about 1980 (the Berke-
ley typology) uses weight to distinguish larger dart points from smaller, but similarly shaped, arrow points. The typology com-
monly used today (the Monitor typology) uses basal width to distinguish wide-based dart pointsfrom narrow-based arrow points.
The two typologies are in general agreement except in central Nevada, where some dart points are light, hence incorrectly typed
by the Berkeley typology, and in eastern California, where some arrow points are wide-based, hence incorrectly typed by the
Monitor typology. Scarce raw materials and resharpening may explain why dart points are sometimes light in central Nevada.
That arrow point basal width is more variable in eastern California than central Nevada likely reflects differences in the cultural
processes attending the spread and subsequent maintenance of bow-and-arrow technology in these two localities.

La disminuci6n en el tamafio de las puntas proyectil hacia 1350 AP se considera generalmente una indicaci6n de la sustituci6n
del atlatlpor el arco yflecha en la Gran Cuenca de los Estados Unidos. La tipologia coman en esta drea antes de 1980 (tipologia
de Berkeley) utiliz6 elpesopara distinguirpuntas mds grandes de puntas mdspequefiaspero deformaparecida. La tipologia gen-
eralmente empleada hoy (Monitor tipologia) utiliza el ancho para distinguir puntas de bases anchas de puntas de bases estrechas.
Tipicamente, estos dos sistemas de tipologia estdn de acuerdo con la excepcion de Nevada central donde algunas puntas proyec-
til ligeras estdn clasificadas equivocamente en la tipologia de Berkeley. Ademds, en California oriental algunas puntas de proyec-
til tienen bases anchas y estdn clasificadas equivocamente en la tipologia Monitor. Probablemente, las puntas de Nevada central
son ligeras debido a lafalta de materia prima y al proceso de reafilaci6n. Hay mds variabilidad en el ancho de las puntas en Cal-
ifornia oriental que en Nevada central y probablamente refleja diferencias en los procesos culturales relacionados a la propa-
gaci6n y mantenimiento subsiguiente de la tecnologia de arco yflecha en estas dos localidades.

ypologies are basic to archaeology. Well-
T defined artifact types facilitate communica-
tion between archaeologists and permit
recognition of regional and temporal patterns that
would otherwise pass unnoticed. In that sense, it can
be argued that typologies themselves are of no intrin-
sic interest; they are merely intermediary construc-
tions useful in investigating the "real" behaviors and
processes we want to study. In this view, typologies
should be evaluated mainly in pragmatic terms:
"good" ones work-they reveal the patterns in which
we are interested; "bad" ones don't. There is much
to be said for this view. Nevertheless, it is occasion-
ally worth asking why our "good" typologies work-
and more importantly, why and where they don't. As
we demonstrate below, such analyses can unexpect-
edly reveal novel patterns and behaviors as impor-
tant and interesting as those for which the typologies
were originally designed. In particular, we think such
work is likely to be especially revealing of basic evo-
lutionary processes connected with the way indi-
viduals acquire, modify, and transmit basic cultural
knowledge. This is so because "good" typologies
identify consistently recurring combinations of
attributes, suggesting the presence of evolutionary
forces that caused these combinations to be main-
tained more or less intact across space and time.
Myriad processes can produce such associations, of
course, but, as we show here, it is possible to narrow
the possibilities by observing the performance level
of different typologies within given units of time and
space. The first part of our analysis is given to such
a comparison, showing how two different projectile
point typologies succeed and fail in two regions of
the western Great Basin of North America. We
demonstrate that these typological successes and fail-
ures are due to regional differences in morphology
and attribute correlation in generally similar point

Robert L. Bettinger * Department of Anthropology, University of California, Davis 95616

Jelmer Eerkens * Department of Anthropology, University of California, Santa Barbara 93106

American Antiquity, 64(2), 1999, pp. 231-242

Copyright ? 1999 by the Society for American Archaeology

231

This content downloaded from 144.82.199.209 on Mon, 16 May 2016 13:33:55 UTC
All use subject to http://about.jstor.org/terms
 232 AMERICAN ANTIQUITY [Vol. 64, No. 2, 1999

types. In the second part of our analysis, we argue

these regional differences in morphology and
attribute correlation are due to differences in the
degree to which dart points were resharpened and to
differences in the cultural mechanisms through
a b ed

which a new technology-the bow and arrow-

spread and was maintained in different parts of the
western Great Basin. The latter argument is informed
by use of a version of evolutionary theory, termed
e f 9 h
culture transmission theory (or dual inheritance the-
ecm
ory). We close with a brief discussion regarding the
relationship of cultural transmission to Darwinian
evolution and of the importance of identifying dif-
ferent modes of cultural transmission in the archae-
ological record. i J ~ ~ ~ ~ ~~~k I m

Typologies in Conflict

The subject of our discussion is an unexpected con-

flict between two "good" Great Basin projectile point
n o P q
typologies. One is the Berkeley typology developed
Figure 1. Corner-notched Projectile points from the
by Robert Heizer and others in the 1960s at the Uni-
White Mountains, California. a-h, Elko Corner-notched.
versity of California-Berkeley (Baumhoff and Byrne i-m, Wide-based Rosegate. n-r, Narrow-based Rosegate.
1959; Clewlow 1967, 1968; Heizer and Baumhoff
1961; Heizer et al. 1968; Heizer and Clewlow 1968; i.e., distinguishing corner-notched points dating
Lanning 1963). The other is the Monitor Valley typol- 3150-1350 B.P. (Elko) from corner-notched points
ogy developed by Thomas (1981). Both of these dating 1350-650 B.P. (Rosegate). The Berkeley
typologies were developed to identify time-sensitive typology uses weight: Rosegate points weigh less
projectile points that could be used in dating archae- than 3 gm; Elko Corner-notched points weigh more
ological sites, especially surface sites that resist dat-
ing by other means, which are especially common in
the Great Basin. The conflict is unexpected because
the Monitor typology is a revision of an earlier typol-
I I

ogy that Thomas (1970) designed specifically to for-

malize the Berkeley typology and duplicate its results / ,

using explicit quantitative criteria. I -~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

The Monitor typology has been immensely suc-

cessful in bringing coherence to Great Basin pro-
jectile point studies, and, overall, it reproduces the
Berkeley typology (Bettinger 1975:167-189). As we
shall show, however, there is systematic disagree-
Central
ment in some parts of the western Great Basin on Nevada

Whtek
the identification of two key forms: Elko Corner- Villages--

notched, a large point form held to date between

Eastern

3150-1350 B.P. (i.e., in the western Great Basin), California

and Rosegate, a smaller corner-notched point form

held to date between 1350-650 B.P. (Figure 1). 'This
I

size difference is commonly regarded as marking the

replacement of the atlatl by the bow and arrow
(Fenenga 1953; Lanning 1963:249). The question Figure 2. Map locating Eastern California and Central
immediately at hand, however, is one of telling time, Nevada.

This content downloaded from 144.82.199.209 on Mon, 16 May 2016 13:33:55 UTC
All use subject to http://about.jstor.org/terms
 Bettinger and Eerkens] BOW-AND-ARROW TECHNOLOGY IN THE GREAT BASIN 233

Table 1. Berkeley and Monitor Valley Classification of Rosegate and Elko Corner-notched Points from Eastern California.

MONITOR

Rosegate Elko Corner-notched Total

Rosegate 113 45 158

BERKELEY Elko Corner-notched 37 37
Total 113 82 195

Note: Data from Ainsworth and Skinner 1994; Basgall and McGuire 1988; Bettinger 1989; Bouscaren 1985; Burton 1986;
Clarke et al. 1991; Delacorte and McGuire 1993: Appendices 1-3; Delacorte et al. 1995: Appendix; Eerkens 1998; and
Gilreath 1995: Appendix.

Table 2. Berkeley and Monitor Valley Classification of Rosegate and Elko Corner-notched Points from Central Nevada.

MONITOR

Rosegate Elko Corner-notched Total

Rosegate 152 122 274

BERKELEY Elko Corner-notched 5 248 253
Total 157 370 527

Note: Data from Thomas 1983:Table 44, 45, 1988:Tables 5, 6, 20, 21, 46,47, 49, 53, 54, 57, 59, 60, 61, 62.

than that (Clewlow 1967; Heizer and Baumhoff
1961; Lanning 1963; O'Connell 1967).Alternatively,
on the premise that resharpening makes weight unsta-
ble (by changing original weight), the Monitor typol-
ogy uses basal width, which is less affected by
resharpening. In the Monitor typology, Elko points
have basal widths greater than 10 mm; Rosegate
points have basal widths less than that (Thomas 1981:
14-15, 19-22).
Because the two typologies use different criteria
to distinguish the age of corner-notched points, some
disagreement is to be expected. In eastern Califor-
nia (Figure 2, Table 1) and central Nevada (Table 2),
however, these disagreements are too large and sys-
tematic to be ignored. Specifically, of the points that
are Rosegate by the Berkeley typology (weight), 28
percent in eastern California and 45 percent in cen-
tral Nevada are Elko Corner-notched by the Moni-
tor typology (basal width). In both places, the
samples in question include many sites and surface
isolates, suggesting this is a regional phenomenon.
It is pointless to argue whether weight or basal width
is inherently better at distinguishing older (Elko)
from younger (Rosegate) corner-notched points,
3
because both methods achieve only local success.
In central Nevada, basal width is superior to
weight. At Gatecliff Shelter, for example, the 10 mm
basal width cutoff correctly predicts the stratigraphic
position of about 97 percent of all corner-notched
points relative to the boundary between Horizons 3
and 4, which dates to roughly 1350 B.P. (Thomas
1983: 177). By contrast, weight is a poorer predic-
tor of age at this site, the 3 gm cutoff correctly pre-
dicting stratigraphic position of only 72 percent of
these points.
In eastern California, on the other hand, weight
is superior to basal width. For instance, as Lanning
(1963: Table 3) showed, the Berkeley typology cor-
rectly predicts the stratigraphic position of Elko and
Rosegate points above or below 48 in (122 cm) 94
percent of the time at the Rose Spring Site (CA-Iny-
372). This near-complete stratigraphic separation by
weight was confirmed by Davis (1963), and again
byYohe (1992: Tables 17, 24). Basal width is much
less predictive of age in comparison. Yohe (1992:
180,Tables 14a-d) measured 145 Rosegate points
recovered from all contexts (i.e., surface and buried)
at the Rose Spring site between 1951 and 1989,
including those in the collections of Lanning and
Davis. Of these, all 53 complete specimens weigh
less than 3 gm, as they should according to the Berke-
ley typology.4 By contrast, basal widths greater than
10 mm cause the Monitor typology to classify 26 (27
percent) of those 96 specimens measurable on this
dimension as Elko Corner-notched. As a result, the
Monitor Valley typology correctly predicts the strati-
graphic position of the 98 Elko and Rosegate points
found in buried contexts and measurable for basal
width only about 67 percent of the time (Yohe 1992:
Tables 14a-d, 17).
Similarly, the larger eastern California sample
demonstrates that weight is superior to basal width

This content downloaded from 144.82.199.209 on Mon, 16 May 2016 13:33:55 UTC
All use subject to http://about.jstor.org/terms
 234 AMERICAN ANTIQUITY [Vol. 64, No. 2, 1999

Tables 3a, b. Weight and Basal Width of Obsidian

points is not a function of their overall size; they are,
Hydration-Dated Rosegate and Elko Corner-Notched Points
on average, shorter, narrower, and lighter than
from Eastern California.
Rosegate points from central Nevada (Table 5).
Weight We summarize the situation in Table 6. In east-

Date < 3 gm 2 3 gm Total

ern California, the Berkeley and Monitor typologies
agree on the identification of corner-notched points
< 1350 B.P. 16 (94%) 1 (6%) 17
older than 1350 B.P., which are consistently both
* 1350 B.P. 5 (28%) 13 (72%) 18
Total 21 14 35 heavy and wide-based, but disagree on the identifi-
cation of points younger than 1350 B.P., which are
Basal Width light (3.0 gm) but often wide-based (10 mm). The

Date < 10 mm >10 mm Total
< 1350 B.P. 25 (66%) 13 (34%) 38
> 1350 B.P. 9 (31%) 20 (69%) 29
Total 34
Berkeley typology correctly classifies these light,
wide-based forms as Rosegate, the Monitor typol-
ogy misclassifies them as Elko. The situation is just
33 67 the reverse in central Nevada, where the typologies

Note: Source specific hydration rates (Hall and Jackson
1989; Basgall and Giambastiani 1995; Delacorte et al. 1995)
were used to derive dates B.P. Many more points can be mea-
sured for basal width than for weight, hence the unequal total
number of observations on the two attributes.
in predicting the age of corner-notched points dated
by obsidian hydration relative to the 1350 B.P.
Elko/Rosegate boundary (Tables 3a,b). Here, the 3
gm cutoff correctly predicts the age 83 percent of the
time, the 10 mm basal width cutoff only 67 percent
of the time. These findings concur with a number of
reports from all parts of eastern California suggest-
ing that corner-notched points younger than 1350
B.P. are consistently light (3 gm) but frequently wide-
based (10mm; e.g. Basgall and Giambastiani 1995:
47, Table B.1; Bettinger 1991a; Delacorte and
McGuire 1993: Appendix A). For example, in alpine
villages in the White Mountains of eastern Califor-
nia, 180 km north of the Rose Springs site (Bettinger
199 la), 36 percent of the Rosegate points occurring
stratigraphically above a tephra layer dating to
approximately 1245 B.P. are wide-based and would
be classified incorrectly as Elko Corner-notched by
the Monitor typology (Table 4; Figure 1 i-m). The
larger basal width of eastern California Rosegate
Table 4. Distribution of Narrow- and Wide-based Rosegate
Points Relative to Ash Layer in Alpine Villages, White
Mountains, California (Bettinger 1991 a).
Narrow-Based Wide-Based
10 mm > 10 mm Total
Above Ash 54 (64%) 31 (36%) 85
In Ash 19 (70%) 8 (30%) 27
(ca. 1245 B.P.)
Below Ash 26 (70%) 11 (30%) 37
Total 99 (66%) 50 (34%) 149
agree on the identification of comer-notched points
younger than 1350 B.P., which are consistently both
light and narrow-based, but disagree on the identifi-
cation of those older than 1350 B.P., which are wide-
based but often light. The Monitor typology correctly
classifies these light, wide-based forms as Elko; the
Berkeley typology misclassifies them as Rosegate.
Explaining Regional Differences in
Point Morphology
What do they mean, these regional differences in
point morphology? Why do Elko Corner-notched
points vary in weight in central Nevada but not east-
ern California, and why do Rosegate points vary in
basal width in eastern California but not central
Nevada? One obvious possibility, of course, is that
Elko Corner-notched points are not all dart points,
as commonly supposed, and that the light Elko Cor-
ner-notched points of central Nevada are really arrow
points that pre-date 1350 B.P. Similarly, it is possi-
ble that not all Rosegate points are arrow points, and
that the wide-based Rosegate points of eastern Cal-
ifornia are really dart points that postdate 1350 B.P.
Metrical data provide little support for these sug-
gestions. Shott (1997) examined a sample of 39
hafted, hence relatively unambiguous, dart points
and 130 similarly unambiguous hafted arrow points,
and used discriminant analysis to derive classifica-
tion functions to separate darts from arrows. Table 7
summarizes the application of Shott's (1997:94) 2-
variable discriminant classification function to the
central Nevada Elko Corner-notched point sample
and the Rosegate points from the Rose Spring site
(CA-Iny-372) in eastern California. These data
strongly suggest that the light Elko Corner-notched
points in central Nevada are dart points, and that the

This content downloaded from 144.82.199.209 on Mon, 16 May 2016 13:33:55 UTC
All use subject to http://about.jstor.org/terms
Bettinger and Eerkens] BOW-AND-ARROW TECHNOLOGY IN THE GREAT BASIN 235

Table 5. Summary Metrical Data for Rosegate Points from Monitor Valley (Thomas 1983, 1988), the Rose Spring Site (Ca-
iny-372; Yohe 1992), and Eastern California (See Table 1 for References). See Thomas (1981, 1983) for
Description of Measurements.

Maximum Axial Maximum Basal Neck

Length Length Width Width Width Thickness Weight
mm mm mm mm mm mm gm

Monitor Valleya
mean 29.68 29.47 16.97 8.19 7.11 3.49 1.64
std 7.30 7.33 3.25 1.22 1.21 0.67 0.68
n 33 33 83 120 120 120 20
Eastern California
mean 25.54 25.05 15.31 8.79 7.78 3.86 1.12
std 5.30 5.57 2.93 2.30 1.67 0.90 0.44
n 55 57 114 157 182 157 39
Rose Spring Site
mean 27.66 no data 15.08 9.10 7.87 3.77 1.17
std 4.91 no data 2.11 1.88 1.46 0.63 0.46
n 83 no data 129 96 130 144 53

a Reference here is to specimens actually measured and omits estimated values routinely given by Thomas (1983, 1988).

wide-based Rosegate points in eastern California are
arrow points.5 Thus, the replacement of Elko Cor-
ner-notched points by Rosegate points about 1350
B.P. in central Nevada and eastern California almost
certainly reflects the replacement of the atlatl by the
bow and arrow, as long suspected.
Unfortunately, this conclusion still leaves us with
our original problem of explaining the weight vari-
ability observed in central Nevada Elko points and
basal width variability observed in eastern Califor-
nia Rosegate points. Following Thomas
(1981:14-15,19-20), we think it likely that exces-
sive resharpening makes dart point weight highly
variable in central Nevada, producing the light, wide-
based Elko Corner-notched points that the Berkeley
typology incorrectly types as Rosegate. Support for
this idea is provided by the length-width ratio; i.e.,
maximum length (maximum width) of light Elko
Corner-notched points in the central Nevada sample,
which is significantly lower than that for the heavy
Elko Corner-notched points in that sample (mean-
light= 1.62, mean = 1.75, p <.001, one-tailed t-
test, df =345), suggesting the stubbier, lighter forms
were more extensively resharpened. That resharp-
ening does not cause similar problems in eastern
California, where Elko Corner-notched points are
uniformly heavy and long relative to width (mean-
length:width =2.02, n =10 complete specimens), may be
explained by the abundance of high-quality obsid-
ian sources there, perhaps causing Elko points to be
discarded without substantial resharpening (cf. Bet-
tinger 1991b; Delacorte 1994). Resharpening, of
course, cannot explain why Rosegate points are more
variable in basal width in eastern California than
central Nevada. We think this variance can be attrib-
uted to differences in how these regions' inhabitants
obtained and subsequently modified bow-and-arrow
technology.
Cultural Transmission and the Spread of the
Bow and Arrow
As noted above, it has long been accepted that the
appearance of Rosegate points marks the advent of
the bow and arrow (Jennings 1986: 116) as reflected

Table 6. Summary of Formal Variation in Elko Corner-notched and Rosegate Points and Success of Berkeley and Monitor
Typologies in Identifying Them in Eastern California and Central Nevada.

Eastern California Typology Central Nevada Typology

Forms Success Forms Success

Elko Corner-notched Heavy, Wide-based Both Correct Heavy, Wide-based Both Correct
Light, Wide-based Berkeley Incorrect

Rosegate Light, Narrow-based Both Correct Light, Narrow-Based Both Correct

Light, Wide-based Monitor Incorrect

This content downloaded from 144.82.199.209 on Mon, 16 May 2016 13:33:55 UTC
All use subject to http://about.jstor.org/terms
 236 AMERICAN ANTIQUITY
Table 7. Classification of Central Nevada Elko Corner-
notched and Eastern California Rosegate Points from the
Rose Spring Site (Ca-iny-372) Using the 2 Variable
Function of Shott (1997).
Dart Arrow Total
Central Nevada
Heavy Elko Corner-notched 97% 3% 229
Light Elko Corner-notched 76% 24% 120
All Elko Corner Notched 90% 10% 349
Shott (1997) Sample Darts 85% 15% 39
Eastern California (CA-INY-372)
Narrow-based Rosegate 5% 95% 66
Wide-based Rosegate 9% 91% 23
All Rosegate 6% 94% 89
Shott (1997) Sample Arrows 11% 89% 130
by a reduction in point size (Fenenga 1953; Lanning
1963). This size reduction has been recognized across
all of the Great Basin, consistently around
1650-1350 B.P. Appearance of the Rosegate series
at essentially the same time throughout the Great
Basin is usually taken to support the companion
assumption (generally unstated) that the spread of
bow-and-arrow technology was a unitary phenome-
non- that is, it was accomplished by the same mode
of cultural transmission across the entire Great Basin,
probably from a common source. Under this assump-
tion, the changes in projectile point morphology from
Elko to Rosegate that accompanied the introduction
of the bow and arrow should be essentially the same
throughout the Great Basin. This is not the case. As
we have shown, the Elko-Rosegate transition in both
central Nevada and eastern California almost cer-
tainly represents the introduction of the bow and
arrow somewhere around 1350 B.P. However, the
characteristics that distinguish Elko from Rosegate
points in central Nevada, and thus the transition to
the bow and arrow, are not the ones that distinguish
them, and thus the transition, in eastern California.
In central Nevada, Rosegate points fit the criteria
of both Berkeley and Monitor typologies: a corner-
notched point with a narrow base is nearly always
light. This suggests that in central Nevada the two
variables these typologies use to distinguish Rosegate
from Elko-weight and basal width-were linked
together in the transmission of bow-and-arrow tech-
nology. Weight and basal width, however, are not
linked in this way in eastern California. This is clearly
illustrated in Table 8, which shows correlation coef-
ficients between weight and basal width in Rosegate
age corner-notched points from multiple sites in east-
This content downloaded from 144.82.199.209 on Mon, 16 May 2016 13:33:55 UTC
All use subject to http://about.jstor.org/terms
[Vol. 64, No. 2, 1999
Table 8: Pearson's R Correlation Coefficients for Basal
Width and Weight in Central Nevada and Eastern California
Rosegate Points.
Monitor Typology Berkeley Typology
Central Nevada .80 .65
Pr=O < .000 1 Pr=O = .0005
Eastern .06 .14
California Pr=O = .36 Pr=O = .20
Note: Correlations are on actual measurements because basal
width and weight estimates for broken points given in
Thomas (1983, 1988) are unavailable for eastern California.
ern California and central Nevada. Using the Mon-
itor typology, basal width and weight are highly cor-
related in Rosegate points from central Nevada.
Because these points nearly always simultaneously
fit the criteria of both typologies, the same high cor-
relation is achieved when they are classified accord-
ing to the Berkeley typology. On the other hand, no
matter which way one classifies them, Rosegate
points from eastern California are uncorrelated on
these two attributes. This suggests that the circum-
stances surrounding the spread and maintenance of
bow-and-arrow technology during Rosegate times in
eastern California were different than those in cen-
tral Nevada.
Boyd and Richerson (1985: 94-95, 243) have
identified two contrasting modes of cultural trans-
mission, guided variation and indirect bias, that are
useful in interpreting these differences in trait corre-
lation and their implications for the spread and main-
tenance of bow-and-arrow technology in the Great
Basin. In guided variation, individuals acquire new
behaviors by directly copying other social models
and subsequently modifying these behaviors to suit
their own needs by individual trial-and-error experi-
ments. Complex behaviors are frequently compiled
in this fashion, using different individuals as social
models for various components of the behavior. The
result is a composite behavior that is more or less
unique to the individual, i.e., as a consequence of
experimentation and the particular set of social mod-
els that were chosen. In indirect bias, on the other
hand, individuals acquire complex behaviors by
choosing a single social model on the basis of a trait
that is deemed to index general proficiency in the
activity to which the desired behavior is related.
Highly successful hunters, for example, might be
chosen as social models by those trying to learn how
to make all sorts of hunting gear. In this case, in con-
 Bettinger and Eerkens] BOW-AND-ARROW TECHNOLOGY IN THE GREAT BASIN 237

Table 9. Pearson's r Correlation Coefficients for Various Point Attributes in Central Nevada and Eastern California Rosegate Points.

Distal Proximal
Axial Maximum Basal Neck Shoulder Shoulder
Length Width Width Width Thickness Angle Angle Weight
Max. C. NV 1.00 .58a .48a .51 .5 la -.19 -.34 .86
Length E. CA 1.00 .24 .07 .21 .15 -.24 -.24 .76
Axial C. NV .59a .49a .5la .49a -.18 -.33 .86
Length E. CA .22 .04 .16 .14 -.24 -.27 .74
Maximum C. NV .44 .72 .12 -.48 -.53a .62
Width E. CA .54 .69 .26 -.47 -.08 .71
Basal C. NV .71 .21 .11 .26 .80a
Width E. CA .78 -.05 -.06 .31 .14
Neck C. NV .23 -.15 -.15 .61
Width E. CA .11 -.06 .15 .40

Thicknes C. NV .14 -.02 .70

Thickness E. CA .03 -.02 .57
Distal C. NV .53a -.12
Shoulder Angle E. CA .16 -.38
Proximal C. NV -.21
Shoulder Angle E. CA -.01

Note: Correlations are on actual measurements. See Thomas (1981, 1983) for description of measurements.
aSignificantly stronger correlation (ou = 0.05).

trast to guided variation, the result is a complex behav-
ior that matches more or less closely in all details of
the behavior of just this one social model. Further,
since there tends to be general agreement within local
groups about the proficiency of potential social mod-
els, the individual generally deemed the most profi-
cient will frequently be chosen as a social model by
many individuals trying to learn new, complex behav-
iors. It is impossible to observe these transmission
processes directly in the archaeological record, of
course, but their statistical signatures should be clear
nonetheless. Variables acquired by guided variation
will be much less strongly correlated than variables
acquired by indirect bias. Following this logic, we
contend that in eastern California, where basal width
and weight are poorly correlated, bow-and-arrow
technology was maintained, and may have spread
initially, by guided variation. Conversely, in central
Nevada, where the attributes are strongly correlated,
we suggest it was maintained, and may have spread
initially, by indirect bias.
In many ways, the situation in eastern California,
where basal width and weight are not linked, is the
one to be expected. This is so because, beyond the
minimal effect basal width has on total weight, these
two attributes respond to different design constraints
that are capable of varying independently, as they
clearly do in eastern California Rosegate points.
However, basal width and weight do not vary inde-
pendently, but are instead correlated in central
Nevada Rosegate points. This correlation suggests
that these elements of design were connected-not
as a matter of function, but because central Nevada
Rosegate point makers acquired the multiple ele-
ments of arrow point design as a package using a
mode of transmission akin to indirect bias, copying,
as it were, the whole point rather than individual
attributes piecemeal and independently. On the other
hand, that weight and basal width are uncorrelated
in eastern California suggests the bow-and-arrow
technology may have spread, and was in any case
maintained, by a mode of cultural transmission akin
to guided variation, in which craftsmen copied, eval-
uated, and modified the various elements of point
design independently. This hypothesis is strongly
supported by observed differences in strength of cor-
relation between other major attributes of Rosegate
points, which are consistently larger for central
Nevada than eastern California (Table 9). In 22 of
these 36 paired correlations, the central Nevada value
exceeds the eastern California value, far more than
the 18 one would expect under the null hypothesis
of no difference in magnitude of attribute correla-
tion in the two samples (p =.015). The central Nevada
correlation is significantly larger in 10 of these cases
7
(p = 0.05). In short, during Rosegate times in cen-
tral Nevada, individuals seem to have maintained
(and may have acquired) this new weapon system-
bow, arrow, and point-as a complete package, while
individuals in eastern California maintained (and

This content downloaded from 144.82.199.209 on Mon, 16 May 2016 13:33:55 UTC
All use subject to http://about.jstor.org/terms
 238 AMERICAN ANTIQUITY
may have acquired) its attributes individually using
a great deal of experimentation.
Discussion
Why the peoples of eastern California might have
acquired and maintained bow-and-arrow technology
through a different mode of transmission than the
peoples of central Nevada is unclear. Following a
suggestion of the late M. A. Baumhoff, however, we
would tentatively propose that eastern California
groups may have acquired the bow and arrow from
peoples with whom they interacted only minimally,
possibly a different linguistic unit occasionally con-
tacted through trade. Owing to this lack of contact,
individuals may have had to perfect a workable bow-
and-arrow technology largely by trial and error. Bet-
tinger (1989: 64-65,229-232) previously noted that
Rose Springs points from eastern California seem
unusually prone to breakage (especially across the
neck) and suggested this faulty design might reflect
early arrow point-makers experimenting with vari-
ous combinations of point size, basal width and fore-
shaft diameter. Perhaps some craftsmen attempted
to adapt existing dart foreshaft types fitted to wide-
based arrow points, while others tried to develop
new, narrower foreshafts fitted to narrow-based arrow
points. Alternatively, craftsmen may have initially
adopted the bow and arrow as a complex, as in cen-
tral Nevada, but immediately set about modifying it
to suit their own individual preferences, so that, as a
result, weight, basal width, and other attributes in the
points came to be relatively uncorrelated. On the
other hand, the spread and perpetuation of bow-and-
arrow technology in central Nevada seems to have
relied more heavily on social transmission, perhaps
facilitated by closer social contacts than character-
ized the situation in eastern California. Whether the
bow and arrow came to central Nevada from the
same source as in eastern California, from an entirely
different area, or perhaps even from eastern Cali-
fornia itself, is unclear. What seems clear, however,
is that point makers in central Nevada adopted the
projectile point used with the bow and arrow by faith-
fully copying all its various attributes in detail, which
suggests they experimented very little with this new
technology, then or subsequently.
Is It Evolution?
The thrust of our paper is about cultural transmis-
sion: how cultural behavior is acquired, modified,
This content downloaded from 144.82.199.209 on Mon, 16 May 2016 13:33:55 UTC
All use subject to http://about.jstor.org/terms
[Vol. 64, No. 2, 1999
and subsequently transmitted, and how one might go
about identifying different kinds of cultural trans-
mission in the archaeological record. Such mecha-
nisms seem to us to be of great potential importance
for humans, because so much of our behavior is
acquired socially rather than determined by individ-
ual learning or genes. Whether this reliance on social
transmission so fundamentally separates humans
from other biological organisms as to require that we
be regarded as unique, and in some sense beyond the
reach of forces that shape the rest of the biological
world, has been endlessly debated inside and outside
anthropology. Certainly, such a disconnection might
be implied if cultural transmission had nothing to do
with genetic fitness. We only say "might" because
Darwinian theory is really not about genes any more
than it is just about the differential reproductive fit-
ness of individuals. Genes are the linchpin in how
biological evolution works but genes as we now
know them are required neither by Darwinian the-
ory nor for it to operate: Darwin got the story basi-
cally right without them; the "modern synthesis"
occurred without Watson and Crick. Darwinian evo-
lution requires a mechanism of transmission and
genes happen to serve this role in biological evolu-
tion. It is possible to imagine all sorts of other trans-
mission mechanisms. Genetic reproduction itself is
not a unitary phenomenon. Asexual and sexual repro-
duction, and in sexual reproduction, the transmission
of sex-linked and autosomal traits, for example, are
quite different and require comparably different
quantitative algorithms. One could not possibly
understand the individual fitness of a biological
organism without also knowing whether it repro-
duced asexually or sexually and, in the latter case,
which traits were sex-linked and whether it was
monogamous or polygamous. The "details" of repro-
duction and mating count. No one model applies
universally. Of course, there is more to biological
evolution than just individuals-populations and
species matter, for example. The genius of Darwin
and especially those after him was in tracing through
complex evolutionary recursions: Forces act on spe-
cific individuals, play out at the population level,
affect individuals in return, and so on. The forces
need not be strong. Minor differences often have
unexpectedly large evolutionary consequences. Here,
again, individual fitness is important in this process,
but there are myriad other, sometimes more impor-
tant, forces. Causal relationships are often complex
Bettinger and Eerkens] BOW-AND-ARROW TECHNOLOGY IN THE GREAT BASIN 239
and evolutionary outcomes are frequently counter-
intuitive.
It seems clear to us that cultural transmission sim-
ply must affect Darwinian fitness-how could it be
otherwise? And Darwinian fitness also must bear on
cultural transmission. Again, how could that not be
true? At minimum, humans must have the biologi-
cal, hence, genetically transmitted, ability for the
cultural transmission of behaviors that certainly
affect Darwinian fitness. It is obvious, at the same
time, that cultural transmission differs in funda-
mental ways from any form of genetic transmission.
The two are asymmetric. Again, this is what we
would expect, since cultural transmission must be
doing something that genes cannot do, just as sex-
ual reproduction is doing something that asexual
reproduction cannot. It does not follow, however,
that this process disconnects cultural transmission
from Darwinian fitness. To the contrary, as with sex-
ual reproduction, the human use of cultural trans-
mission is simply the exploiting of an evolutionary
opportunity. To deny that would imply that the cul-
turally-mediated evolutionary success of anatomi-
cally modern humans is merely serendipitous
happenstance.
Do Differences in Cultural
Transmission Matter?
There is general tendency for archaeologists to
assume that differences in cultural transmission are
unimportant (Bettinger et al. 1996). We suppose that
reflects a common misperception that Darwinian
forces are all obvious, strong, and life-threatening.
That, of course, is to misread Darwin, whose uni-
formitarian gradualism stood in direct opposition to
explanations relying on supremely powerful forces
and catastrophes capable of instantaneously trans-
forming the world and its species. As we have said,
in Darwinian evolution details often matter, and that
is the case here.
Our paper contrasts two different modes of cul-
tural transmission, guided variation and indirect bias,
that highlight what is perhaps the fundamental con-
trast in evolutionary potential between genetic and
cultural transmission. As Boyd and Richerson
(1985:132-171; see also Cavalli-Sforza 1988)
observe, where cultural transmission takes the form
of guided variation and other modes of transmission
involving substantial individual experimentation and
learning, human behavior will tend to optimize fit-
This content downloaded from 144.82.199.209 on Mon, 16 May 2016 13:33:55 UTC
All use subject to http://about.jstor.org/terms
ness in accord with the predictions of the genetic
model. By contrast, where cultural transmission is
by indirect bias and other modes that bypass indi-
vidual experimentation and learning, there is an
opportunity for a much different range of behaviors
that are normally precluded when only genes are
involved. This situation largely occurs because indi-
rect bias and related forms of social transmission tend
to produce behaviorally homogenous local popula-
tions. It is precisely under these conditions that the
force of selection can fall more heavily on groups
than on individuals, i.e., as group selection. That, in
turn, makes it easier to understand and explain a host
of human characteristics suggesting the presence of
selective forces acting at the group level-most
notably our ultra-social character and tendency to
cooperate despite sometimes extensive personal
costs-that have proven difficult to explain with ref-
erence to the genetic model.
It has been argued that the hunting behavior of
certain hunter-gatherer groups contains built-in
restraints that act to prevent resource depletion (e.g.,
Moore 1965). Such an explanation, however, does
not account for individual hunters devising and using
alternative behaviors that produce higher short-term
returns and thus a benefit from the resource abun-
dance arising from the more restrained practices of
their fellow hunters. In that event, restraint quickly
gives way to strategies that are more successful in
the short term, as in the familiar "tragedy of the com-
mons" (Hardin 1968). In short, individual learning
and experimentation prevent the development and
maintenance of behaviors that potentially benefit the
group. Conversely, because it insulates cultural trans-
mission from both individual learning/experiment
and exotic social models, indirect bias produces the
conditions under which group-beneficial behaviors
can evolve and persist. Accordingly, if, during
Rosegate times in eastern California, behaviors con-
nected with hunting (e.g., when, where, and how to
hunt, and who to hunt with) were acquired and trans-
mitted by the same means as bow-and-arrow tech-
nology, the emphasis on individual learning and
experimentation likely would have prevented the
development of group-beneficial cooperative behav-
iors. By contrast, the emphasis on indirectly biased
social transmission and imitation suggested by the
uniformity of Rosegate projectile point technology
in central Nevada implies the presence of the ideal
conditions for such group-beneficial, cooperative
 240 AMERICAN ANTIQUITY
behaviors to develop and persist. Thus, despite the
general similarities in technology, it is quite possi-
ble that hunting behaviors and social organization
overall differed substantially between the two regions
during this interval.
Supporting this argument is evidence which sug-
gests that Numic-speaking groups occupying the
Great Basin in ethnographic times spread rapidly out
of eastern California sometime after 1000 B.P. (Bet-
tinger and Baumhoff 1982; see also Madsen and
Rhode 1994). This rapid occupation, it is argued, suc-
ceeded through the competitive advantages of the
Numic adaptive strategy over that of pre-Numic peo-
ples, and the latter's failure to readapt to Numic com-
petition through slow culture change partly caused
by indirectly biased social transmission (although not
specifically labeled as such; Bettinger and Baumhoff
1982: 488-493). The emphasis on indirectly biased
social transmission suggested here for central Nevada
during Rosegate times is clearly consistent with this
hypothesis.
Our hypothesis is just that-an hypothesis. Yet it
remains that weight, basal width, and nearly all other
attributes are highly correlated in Rosegate points
from central Nevada and poorly correlated in
Rosegate points from eastern California. The lack of
correlation in eastern California establishes that the
high correlation in central Nevada is unlikely due to
functional constraints. Accordingly, we have chosen
to interpret these differences in correlation using the
tenets of cultural transmission theory, on the hypoth-
esis that attributes passed on through processes that
emphasize social learning over individual learning,
such as indirectly biased transmission, should be
more highly correlated than those passed on through
processes that emphasize individual over social
learning, such as guided variation.
It is widely held that evolutionary theory, espe-
cially culture transmission theory, has little to offer
the archaeologist, partly because evolutionary
processes are difficult to detect in the archaeologi-
cal record. We have shown, with reasonable samples,
the possibility for casting standard archaeological
data in ways that reveal the basic mechanisms
through which cultural evolution operates-in this
case, the way in which real individuals acquired and
subsequently modified important cultural knowl-
edge. Our main point is that differences in cultural
transmission are detectable archaeologically and, in
fact, seem to distinguish the maintenance and pos-
This content downloaded from 144.82.199.209 on Mon, 16 May 2016 13:33:55 UTC
All use subject to http://about.jstor.org/terms
[Vol. 64, No. 2, 1999
sibly introduction of bow-and-arrow technology in
different parts of the Great Basin. It is about evolu-
tionary theory, then, to the extent that transmission,
in this case cultural transmission, is a critical Dar-
winian process that requires investigation. The
approach is straightforward and applicable in a vari-
ety of other archaeological contexts. We hope our
application will stimulate further research and debate
into the study of evolutionary culture change in the
Great Basin and elsewhere.
Acknowledgments. We are indebted to Elizabeth Klarich,
who provided the Spanish abstract translation; Robert Yohe,
who provided data for points from the Rose Spring site; and
Susan Harriss, Cynthia Herhahn, Mike Jochim, Lisa Deitz,
Frank Deitz, and Brian Ramos who read and commented on
various drafts of this paper and helped out here and there
with the manipulation of data. We further wish to thank
Lynne Goldstein, editor of American Antiquity, and three
serious reviewers who commented on the previous version of
this paper.
References Cited
Ainsworth, P., and E. Skinner
1994 Appendix 1. In ArchaeologicalResour ces Evalluation at
CA-INY-384, CA-INY-3790, CA-INY-4547, CA-INY-4549H,
CA-INY-4550, Inyo County California, edited by B. Wick-
strom, R. Jackson, and T. L. Jackson, pp A1-1-89. Report
submitted to California Department of Transportation, Envi-
ronmental Branch, District 9, Bishop, California. Contract
09H077.
Basgall, M. E., and K. R. McGuire
1988 The Archaeology of CA-INY-30: Prehistoric Culture
Change in the Southern Owens Valley, California. Report
submitted to California Department of Transportation, Sacra-
mento. Contract No. 09G519. Far Western Anthropological
Research Group, Davis, California.
Basgall, M.E., and M.A. Giambastiani
1995 Prehistoric Use of a Marginal Environment: Continuity
and Change in Occupation of the Volcanic Tablelands, Mono
and Inyo Counties, California. Publication 12. Center for
Archaeological Research, Davis, California.
Baumhoff, M.A., and J.S. Byrne
1959 Desert Side-Notched Points as a Time Marker in Cali-
fornia. Archaeological Survey Reports 48: 32-65. Univer-
sity of California, Berkeley.
Bettinger, R.L.
1975 The SuifaceArchaeology of Owens Valley, Easter n Cal-
ifornia: Prehistoric Man-Land Relationships in the Great
Basin. Unpublished Ph.D. dissertation, Department of
Anthropology, University of California, Riverside.
1989 The Archaeology of Pinyon House, Two Eagles, and
Crater Middens: Three Residential Sites in Owens Valley,
Eastern California. Anthropological Papers 67. American
Museum of Natural History, New York.
1991a Aboriginal Occupation at HighAltitude: AlpineVillages
in the White Mountains of Eastern California. American
Anthropologist 93: 656-679.
1991b Native Land Use: Archaeology and Anthropology. In
Natural History of the White-Inyo Range, Eastern Califor-
nia, edited by C. A. Hall, pp. 463-486. University of Cali-
fornia Press, Berkeley.
 Bettinger and Eerkens] BOW-AND-ARROW TECHNOLOGY IN THE GREAT BASIN 241
Bettinger, R. L., and M. A. Baumhoff
1982 The Numic Spread: Great Basin Cultures in Competi-
tion. American Antiquity 47:485-503.
Bettinger, R. L., P. J. Richerson, and R. Boyd
1996 Style, Function, and Cultural Evolutionary Processes. In
Darwinian Archaeologies, edited by D.G. Maschner, pp.
133-164. Plenum Press, New York.
Bettinger, R.L., and J.W. Eerkens
1997 Evolutionary Implications of Metrical Variation in Great
Basin Projectile Points. In Rediscover ing Darwhin. Evolu-
tionaiy Theoiy and Archaeological Explanation, edited by
C. M. Barton and G. A. Clark, pp. 177-191. Archaeological
Papers 7. American Anthropological Association, Arling-
ton, Virginia.
Bouscaren, S.
1985 Ar-chaeological Excavations in the Lowlands of Nor-th-
ern Owens Valley: Report on the Sawmill Road Site (CA-
INY-1386). Unpublished Master's thesis. University of
California, Riverside.
Boyd, R., and Richerson, P. J.
1985 Culture and the Evolutionaqy Process. University of
Chicago Press, Chicago.
Burton, J.F.
1986 Ar-chaeologicalInvestigations atBajada Camp (CA-Iny-
2596), Inyo County California. Prepared for Baxter Ranch
by Trans-Sierran Archaeological Research, Columbia, Cal-
ifornia.
Cavalli-Sforza, L.
1988 Cultural Transmission and Adaptation. International
Social Science Journal 40:239-253.
Clark, M., T. Jones, T. Fung, and S. Grantham
1991 Extended Phase I Investigation of CA-INY-3694, CA-
INY-3695, CA -INY-36961H, and CA-INY-36971H, NearLone
Pine, Inyo County, California. Prepared for California
Department of Transportation, District 9, Bishop, Califor-
nia. Contract Number 09-INY-395, P.M 54.6/59.0, 09-
213000.
Clewlow, C. W. Jr.
1967 Time and Space Relations of Some Great Basin Projec-
tile Point Types. Archaeological Survey Reports 70: 141-150.
University of California, Berkeley.
1968 SurfaceArchaeology ofthe BlackRockDesert, Nevada.
Ar-chaeological Survey Reports 73: 1-94. University of Cal-
ifornia, Berkeley.
Davis, J. T.
1963 Test Excavations at Site lny-372 Conducted in 1961. In
Archaeology of the Rose Spr-ing Site Iny-372, by E. Lanning,
Appendix 5, pp. 296-304. Publications inAmericanArchae-
ology and Ethnology 49(3): 237-336. University of Cali-
fornia, Berkeley.
Delacorte, M. G.
1994 The Role of Population in Relation to the Use of Alpine
Environments in the White Mountains of California and
Nevada. Aieta Ar-chaeological Carpathia. Warsaw. In press.
Delacorte, M. G., M. C. Hall, and M. E. Basgall
1995 Final Repor-t on the Evaluation of Twelve Archaeologi-
cal Sites in the Southern Owens Valley, Inyo County, Cali-
fornia. Report submitted to California Department of
Transportation, Sacramento. Contract No. 09HO78. FarWest-
ern Anthropological Research Group, Davis, California.
Delacorte, M. G., and K. R. McGuire
1993 Report of Archaeological Test Evaluations at Twenty-
Three Sites in Owens Valley, California. Report prepared for
Contel of California and the Bureau of Land Management,
California Desert District. Contract No. SC-33-90. FarWest-
ern Anthropological Research Group, Davis, California.
This content downloaded from 144.82.199.209 on Mon, 16 May 2016 13:33:55 UTC
All use subject to http://about.jstor.org/terms
Eerkens, J.W., and R.L. Bettinger
1994 Variance in Great Basin Projectile Points. Paper pre-
sented at the 59th Annual Meeting of the Society for Amer-
ican Archaeology, Anaheim.
Eerkens, J.W.
1998 Field Notes from Archaeological Survery and Excava-
tion in Southern Owens Valley, California. Manuscript on
file, Department of Anthropology, University of California-
Santa Barbara.
Fenenga, F.
1953 The Weights of Chipped Stone Points: A Clue to their
Functions. Southwest Journal of Anthropology 9: 309-323.
Gilreath, A. J.
1995 Archaeological Evaluation of Thirteen Sites for the Ash
Creek Project, Inyo County, California. Report submitted to
California Department of Transportation, Sacramento. Far
Western Anthropological Research Group, Davis, Califor-
nia.
Hardin, R.
1968 Tragedy of the Commons. Science 162:1243-1248.
Hall, M.C., and R.J. Jackson
1989 Obsidian Hydration Rates in California. In Current
Directions in California Obsidian Studies, edited by R.E.
Hughes, pp. 31-58. Contributions No. 48. Archaeological
Research Facility, University of California, Berkeley.
Heizer, R.F., and M.A. Baumhoff
1961 TheArchaeology of Wagon Jack Shelter. In TheArchae-
ology of Two Sites at Eastgate, Churchill County, Nevada.
Anthropological Records 20(4): 119-138. University of Cal-
ifornia, Berkeley.
Heizer, R.F., M.A. Baumhoff, and C.W. Clewlow Jr.
1968 Archaeology of South Fork Shelter (NV-El- 11), Elko
County, Nevada. Archaeological Survey Reports 71: 1-58.
University of California, Berkeley.
Heizer, R.F., and C.W. Clewlow Jr.
1968 Projectile Points from Site NV-CH-15, Churchill County,
Nevada. Archaeological Survey Reports 71: 59-88. Univer-
sity of California, Berkeley.
Jennings, J.D.
1986 Prehistory: Introduction. In Great Basin, edited by W.
L. D'Azevedo, pp. 113-119. Handbook of North American
Indians, vol. 11, W. C. Sturtevant, general editor. Smithson-
ian Institution, Washington, D.C.
Lanning, E.P.
1963 Archaeology of the Rose Spring Site Iny-372. Publica-
tions in American Archaeology and Ethnology 49(3):
237-336. University of California, Berkeley.
Madsen, D. B., and D. R. Rhode (editors)
1994 Across the West: Human Population Movement and the
Expansion of the Numna. University of Utah Press, Salt Lake
City.
Moore, 0. K.
1965 Divination: A New Perspective. American Anthropolo-
gist 59:69-74.
O'Connell, J. F.
1967 Elko Eared/Elko Corner-notched Projectile Points as
Time Markers in the Great Basin. Archaeological Survey
Reports 70: 129-140. University of California, Berkeley.
Shott, M. J.
1997 Stones and Shafts Redux: The Metric Discrimination of
Chipped Stone Dart and Arrow Points. American Antiquity
62:86-101.
Thomas, D. H.
1970 Archaeology'sOperationalImperative:GreatBasinPro-
jectile Points as aTest Case. University ofCaliforniaArchae-
ological Survey Annual Report 12: 27-60. Los Angeles.
 242 AMERICAN ANTIQUITY [Vol. 64, No. 2, 1999

1981 How to Classify the Projectile Points from MonitorVal- 5As one would expect, light Elko Corner-notched points are
ley, Nevada. Journal of California and Great Basin Anthro-
more frequently classified as arrow points than heavy ones.
pology 3: 7-43.
However, the error-rate for the central Nevada Elko Corner-
1983 The Archaeology of Monitor Valley 2. Gatecliff Shelter.
notched sample as a whole (i.e., including light and heavy
Anthropological Papers 59(1). American Museum of Nat-
ural History, New York.
forms) is lower than for Shott's (1997) dart sample, and the

1988 The Archaeology of Monitor Valley 3. Survey andAddi-
tional Excavations. Anthropological Papers 66(2). Ameri-
can Museum of Natural History, New York.
Yohe, R. M. II
1992 Reevaluation of Western Great Basin Cultural Chronol-
ogy and Evidence for the Timing of the Introduction of the
Bow and Arrow to Eastern California Based on New Exca-
vations at the Rose Spring Site (CA-INY-372). Unpublished
Ph.D. dissertation, Department of Anthropology, University
of California, Riverside.
error-rate for light Elko Corner-notched points does not differ
significantly from that obtained in Shott's dart sample. Filtering
Shott's dart sample to include only specimens weighing (3 gm
would almost certainly increase the relative frequency of incor-
rect classifications in much the same way that filtering the Elko
Corner-notched for weight does in the central Nevada sample.
6This is not a function of excessive variation in basal width in
Rosegate points from eastern California. We have shown else-
where (Eerkens and Bettinger 1994; Bettinger and Eerkens
1997) that Rosegate basal width is relatively stable in this

Notes region, and more stable than Rosegate basal width in central
Nevada. Further, those data demonstrate that the variability in
1 The Berkeley and Monitor typologies differ somewhat in ter-
basal width of other point types is essentially the same in east-
minology. The Berkeley typology recognizes separate Rose
ern California and central Nevada (Bettinger and Eerkens 1997:
Spring series and Eastgate series, each with several distinct
Table 10.5), suggesting that raw material access (i.e., to stone,
types. The Monitor lumps all the corner-notched forms of Rose
wood, etc.) is likely not causing the regional difference in
Spring and Eastgate into a single Rosegate series. Because our
Rosegate basal width. The Rosegate series does subsume two
problem concerns corner-notched point forms, we follow the
formally distinct types, Rose Spring Corner-notched and
Monitor convention for the sake of clarity, cautioning the reader
Eastgate Split-stem, but in eastern California, the two are virtu-
that many of the references we cite follow the Berkeley termi-
ally identical in basal width. In the White Mountains sample
nology.
(Bettinger 1991a) illustrates this nicely. Eastgate basal width:
2 In the original version of what became the Monitor typology,
mean = 1.01 cm, std = 0.19, n = 20; Rose Spring Comer-
Thomas (1970) used weight, as in the Berkeley typology.
notched basal width: mean = 0.96, std = 0.18, n = 37).
3 Note, however, that Thomas (1981) did not argue for the uni- 7Note in contrast here, that while eastern California Elko
versal applicability of the Monitor typology.
Corner-notched points tend to be somewhat more highly corre-
4 A fragmentary surface specimen identified by Yohe as lated across major attributes than central Nevada Elko Corner-
Rosegate weighed 3.2 gm (1992: 288, Table 14a). However, this
notched points, the difference between the two is not significant
piece could as easily be classified as Elko. Note also that Yohe
(p = .37).
(1992) inadvertently ommitted metrics for the 30 Rosegate
points he recovered between 1987-1989, but provided these Received December 18, 1997; accepted August 25, 1998;
data (labeled Table 16 ) at our request. revised October 30, 1998

This content downloaded from 144.82.199.209 on Mon, 16 May 2016 13:33:55 UTC
All use subject to http://about.jstor.org/terms